> For the complete documentation index, see [llms.txt](https://docs.tillforty.com/oaura/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.tillforty.com/oaura/readme.md).

# Welcome

Build AI agents that talk to your customers, use your tools and data, and hand off to your team when they need to.

Oaura lets you build an AI agent in the dashboard and put it in front of your customers. Add it to your website with one `<script>` tag, or call it from your own backend through the REST API. The agent answers from your knowledge base, calls your tools (MCP servers, HTTP endpoints, n8n workflows and ready-made integrations), knows who it's talking to, and passes the conversation to a person on your team in Slack when a visitor asks for one.

These docs are for the people building on Oaura: operators who set up agents, and developers who embed, integrate and extend them.

## Where to start

| I want to…                                           | Start here                                                                                                                     |
| ---------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
| Get an agent live on my site in five minutes         | [Quickstart](/oaura/get-started/quickstart.md)                                                                                 |
| Put the chat widget on my website                    | [Embedding the widget](/oaura/put-it-on-your-site/widget/embedding.md)                                                         |
| Talk to an agent from my own backend                 | [Your first API call](/oaura/get-started/your-first-api-call.md), then [REST API overview](/oaura/rest-api/overview.md)        |
| Give my agent tools from an MCP server               | [Connect an MCP server](/oaura/get-started/connect-your-first-mcp-server.md), then [MCP overview](/oaura/connect-tools/mcp.md) |
| Let the agent know which of my users it's talking to | [Identity overview](/oaura/put-it-on-your-site/identity.md)                                                                    |
| Hand conversations to my team                        | [Human handoff](/oaura/put-it-on-your-site/human-handoff.md)                                                                   |
| Answer from my own documents                         | [Knowledge bases](/oaura/build/agents/knowledge-bases.md)                                                                      |

New to Oaura? Read [What is Oaura](/oaura/readme/what-is-oaura.md) and [Key concepts](/oaura/readme/key-concepts.md) first. They take a few minutes and make the rest of the docs easier to follow.

## What you can build

* **Agents** with their own prompt, model and greeting. Run them on Oaura's built-in model, or on your own OpenAI or Anthropic key. See [Agents & models](/oaura/build/agents/agents-and-models.md).
* **A chat widget** for your site, with your colours, starter prompts, voice input and file upload. See [Embedding the widget](/oaura/put-it-on-your-site/widget/embedding.md) and the [configuration reference](/oaura/put-it-on-your-site/widget/configuration-reference.md).
* **Tools** the agent can call: [MCP servers](/oaura/connect-tools/mcp.md), [HTTP endpoints](/oaura/connect-tools/webhooks/http-connector.md), [n8n workflows](/oaura/connect-tools/webhooks/n8n-connector.md) and [integrations](/oaura/connect-tools/integrations-catalog.md).
* **Knowledge bases** the agent searches when a question needs your documents. See [Knowledge bases](/oaura/build/agents/knowledge-bases.md).
* **Verified end users.** Identify signed-in visitors with your own login, and pass who they are to your tools as a signed token. See [Identity](/oaura/put-it-on-your-site/identity.md).
* **Environments** that change the agent's behaviour per customer or audience. See [Environments](/oaura/build/environments.md).
* **Tabs** inside the widget that show your own data, like orders or tickets. See [Tabs](/oaura/build/tabs.md).
* **Human handoff** to your team in [Slack](/oaura/put-it-on-your-site/human-handoff/slack.md): the conversation continues in a Slack thread until either side hands it back to the AI. See [Human handoff](/oaura/put-it-on-your-site/human-handoff.md).
* **A REST API** for conversations, messages and knowledge bases. See the [REST API overview](/oaura/rest-api/overview.md) and the [full API reference](https://api.oaura.app/docs).

## Plans

Oaura has three plans: **Starter**, **Business** and **Enterprise**. Your first workspace starts on a 30-day Business trial. A workspace without a plan is paused until you pick one. Some features, such as identity providers and more than one environment per agent, need Business or above. Pages for plan-gated features say so at the top. See [Plans & limits](/oaura/manage/plans-and-limits.md) for the full comparison.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.tillforty.com/oaura/readme.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.